Your current BG is a snapshot. The trend arrow is a forecast. Together they tell you where glucose is actually headed — which is the number that matters once you start doing anything about it.
Enter your current BG and pick the trend arrow you're seeing. The tool projects where glucose is likely to be in 30 minutes if the trend holds — the window where rapid-acting insulin starts doing real work. This is a learning tool, not a prescription.
Your CGM shows one number. But that number is a photograph — it's where glucose was when the sensor last read. If the arrow is moving, glucose is already somewhere else.
The 30•60•90 Rule is a way to do the simple math in your head: where will BG actually be in 30 minutes? That's the window that matters, because that's roughly when a bolus starts doing real work. Treating a meal based on what the screen says right now is like aiming at where a moving car was, not where it's headed.
Why 30, 60, and 90?
The three numbers are just the three speeds a trend arrow can travel at, projected forward 30 minutes. That's all the name means.
The full chart:
| Arrow | What It Means | Speed | In 30 min, BG moves |
|---|---|---|---|
| ↑↑ | Very fast rise | >3 mg/dL / min | +90 or more |
| ↑ | Fast rise | 2–3 mg/dL / min | +60 to +90 |
| ↗ | Slow rise | 1–2 mg/dL / min | +30 to +60 |
| → | Steady | <1 mg/dL / min | Roughly unchanged |
| ↘ | Slow fall | 1–2 mg/dL / min | −30 to −60 |
| ↓ | Fast fall | 2–3 mg/dL / min | −60 to −90 |
| ↓↓ | Very fast fall | >3 mg/dL / min | −90 or more |
By the time you see the number, glucose has already been moving for a few minutes. By the time you bolus and the insulin starts working, more time has passed. The 30-minute forecast is trying to meet glucose where it actually is, not where it was.
Rapid-acting insulin needs about 30 minutes to really start working. That's why the forecast uses that window — anything shorter isn't actionable, anything longer stops being predictable.
A 120 with a flat arrow means 120 in 30 minutes — nothing to worry about. A 120 with ↑↑ means 210 in 30 minutes — a completely different situation. The forecast is the thing that separates those two moments.
This consumer-friendly formulation is from a Texas Children's Hospital Clinical Nutrition Services handout. The underlying concept — projecting current BG forward 30 minutes before dosing — traces to published clinical frameworks from Pettus & Edelman (2017) and Scheiner.
The 30-minute forecast is reliable under specific conditions. Outside those conditions, the math gets dangerous. These caveats come from the same clinical literature the rule is based on — pay attention to them.
The Endocrine Society specifically warns against using trend arrows to make correction decisions in the hours right after eating, because you can "stack" insulin on top of a bolus that hasn't finished working. This is the biggest safety pitfall.
The 30-minute projection assumes standard rapid-acting insulin (Humalog, Novolog, Admelog). Ultra-rapid insulins (Fiasp, Lyumjev) and inhaled insulin (Afrezza) peak significantly faster — applying these adjustments to them can be dangerously aggressive.
The rate-of-change brackets in this rule (1–2, 2–3, >3 mg/dL/min) match Dexcom-style arrows. FreeStyle Libre, Medtronic Guardian, and other systems define their arrows slightly differently. Check your device's specific definitions.
Trend-arrow projection is considered reliable out to about 30 minutes. Past an hour, it becomes much less predictable. The rule is a short-window tool, not a long-window forecast.
This is a forecast, not a dose. The rule tells you where glucose is headed. It doesn't tell you how much insulin to give, it doesn't tell you when to give it, and it doesn't know anything about what you're about to eat. Fat, protein, fiber, meal size, and timing all affect how a meal actually plays out.
What the forecast does is change the question. You stop asking "I'm at 120, what do I do?" and start asking "I'm heading toward 210 in half an hour — what does that change?" That shift in what you're aiming at is the whole point.
Pair it with the Rule of 10 for timing. Pair it with trend-arrow reading for the full picture. No single rule is the answer.
Scott and Jenny have covered pre-bolus timing, trend arrows, and how to read what's coming across many episodes. Start here.
Pre-Bolusing: The Juicebox Way →